PixelFlow: Pixel-Space Generative Models with Flow
By: Shoufa Chen , Chongjian Ge , Shilong Zhang and more
Potential Business Impact:
Makes computers create amazing pictures from words.
We present PixelFlow, a family of image generation models that operate directly in the raw pixel space, in contrast to the predominant latent-space models. This approach simplifies the image generation process by eliminating the need for a pre-trained Variational Autoencoder (VAE) and enabling the whole model end-to-end trainable. Through efficient cascade flow modeling, PixelFlow achieves affordable computation cost in pixel space. It achieves an FID of 1.98 on 256$\times$256 ImageNet class-conditional image generation benchmark. The qualitative text-to-image results demonstrate that PixelFlow excels in image quality, artistry, and semantic control. We hope this new paradigm will inspire and open up new opportunities for next-generation visual generation models. Code and models are available at https://github.com/ShoufaChen/PixelFlow.
Similar Papers
One-step Latent-free Image Generation with Pixel Mean Flows
CV and Pattern Recognition
Makes computers draw pictures faster and better.
PixNerd: Pixel Neural Field Diffusion
CV and Pattern Recognition
Makes AI create clearer, faster pictures.
Advancing End-to-End Pixel Space Generative Modeling via Self-supervised Pre-training
CV and Pattern Recognition
Makes computers draw clearer, faster pictures.